Export diversification of agricultural products has been a goal for many countries, especially those in the developing world which often face difficulties in moving into higher value-added product categories. The existing body of literature has highlighted the importance of diversification with regards to the sectoral composition  of exports, (Bonaglia and Fukasaka, 2003; Cadot and Strauss-Kahn, 2007, 2012), and regarding the processing and value-added to export products (Taylor and Francis, 2003). Prior work has thus established a connection between economic growth and diversification of both exports and imports (Parteka and Tamberri, 2013).

Import diversification has also been identified as a function of economic growth, and with specific regards to agricultural products (Choudhury and Headey, 2017; Jaimovich, 2012), a trend which in large measure is fueled by a consumption shift towards dietary diversification at certain income thresholds (Regmi, 2001). Geographical diversification has also been highlighted as possibly even more important than product diversification, because widening the scope of export destinations can lead to a rise in demand for additional products from the same export partner (Amurgo-Pecheco and Piérola, 2007). Previous studies have thus identified a general trend of geographical diversification of BRICS country exports and imports (Belardo, 2018), as well as a highly elastic import demand (Haq Zahoor and Meilke, 2009). This appears to have been a parallel trend to the structural rise of developing countries as global food importers (Serrano and Pinilla, 2011). Previous studies have thus examined trade diversification with focus on either specific countries or groups of countries, and most frequently also, analyzing only exports or imports, or only geographical or sectorial diversification. With this contribution, we seek to contribute to this field of study by presenting an unprecedently wide array of data comprising countries worldwide classified according to their main trade groupings for purpose of clarity. Moreover, we provide data on both exports and imports, as well as on both sectorial and geographical diversifications in each of them.

Over the last two decades, agricultural trade has changed as a result of greater trade liberalization brought about by the results of the Uruguay Round and by an increase in number of the Regional Trade Agreements (RTA). Figure 1 shows that, since 2000, the number of trade agreements involving products (not only from the agribusiness sector) has grown across the globe. But has this also been associated with an increase in the diversification of partners and, or, products? Or did these agreements mainly serve to lock in trade patterns between their members? And how do they perform in relation to the diversification of agribusiness trade? This article does not seek to demonstrate a causal relationship between these phenomena, but rather to contribute to this field of study with estimates and analyses of diversification trends, in order to provide a broader perspective upon the evolution of international agribusiness trade in relation to the growth of these agreements.

In this study, we analyze the trade diversification of the imports and exports of individual countries and groups of countries in terms of both products and markets by assessing the degree of sectorial and geographical concentration of their trade flows. We specifically analyze the period from 2000 to 2019. By emphasizing key data and general trends within global agricultural trade, we thereby seek to contribute to analyses of the current and future commercial and political developments within this field. For purposes of space, we mainly focus on the largest players within the field of agriculture, and treat the wider universe of countries in an aggregated manner by analyzing different WTO groups. However, the data for all countries are available in the Annex B. We hereby seek to understand whether they display similar or differing trade patterns and whether changes have been evident throughout the period examined. This information can be useful for policy makers in different countries to obtain a broader perspective of structural changes in trade flows, the differences between countries, and their evolution over time.

To identify the general patterns within the global agricultural trade, we initially analyze the relation between the exported and imported value of agribusiness. We hereafter estimate indexes of diversification: of each country’s agricultural trade partners regarding both imports and exports, and also concerning the specific products exported and imported. 

Our point of departure is therefore the groups formed during the agricultural negotiations in the WTO: the Cairns Group, a coalition of agricultural exporters favoring trade liberalization; the G-10, which is a coalition of countries working for agriculture to be treated as a special field due to non-trade concerns; and the G33, also known as the “Friends of Special Products” within agriculture, a coalition of developing countries that favor flexibility to pursue limited market opening within agriculture. As we sought to conduct a broad global analysis of agricultural trade patterns and many countries did not appear within these groups, we also included other groups, such as the G-90 and the Friends of Ambition for Non-Agricultural Market Access (NAMA), as well as the countries without group. The first is a general negotiation group, comprising various countries not engaged within agricultural negotiations. This group is formed by states in Africa, Africa, Caribbean, and the Pacific Group of States (ACP) and least-developed countries. As many countries appear in more than one group, we considered the group with the largest number of participants. The Friends of Ambition (NAMA) is not specifically concerned with agricultural products, but was included as it comprises of important players within agricultural trade which do not appear within the previously mentioned groups, - the United States and the European Union. We therefore refer to this group as USA & EU. It generally aims towards global market opening with exceptions for agriculture. We also included the countries without group, in order to assess any possible general traits, which might become visible despite its highly heterogeneous composition.

Diversification indexes

The indexes of trade diversification were created to examine  four different dimensions related to: export destinations; export products; import origins; and import products. When trade partners are analyzed (destination or origin in the case of exports or imports, respectively) diversification refers to the geographical dimension of the index. The sectorial dimension relates to trade diversification in terms of products (exports or imports). The agricultural products used and their correspondence with the harmonized system (HS) is described in Table 1A in Annex A. They follow the classification of agricultural products described in United States (2021).         

The diversification index used in this study was the Global Geographic Diversification Index (GGDI) with regards to the geographical diversification, and the Global Sectoral Diversification Index (GSDI) regarding sectorial diversification. Lombaerde and Iapadre (2012) and Bouët and Odjo (2019) describe these indexes, which differ from the traditional index derived from the Herfindahl Concentration Index (HHI) as within the latter, only the sum of the share within each market is measured. Yet, within the GGDI and GSDI, the difference between the share in each market and in the world is considered. The Equations 1 and 2, respectively, describe the GGDI and GSDI indexes used in this study, for a given country (i), considering the dimension of the exports. Therefore, these same indexes were calculated for imports, but considering the imported values.

Where  corresponds to the value of agricultural exports in country i for the country j;  is the global exported value of the agricultural products to country   j;  the exported value of Agricultural products of country i to the world, and  the value of global exports (w) of agricultural products. The  e  are, respectively, the exported values of i and the world for product k. In this case, the products did not correspond to HS, but to a grouping of them in products as presented in Table 1A of the Annex A.

Data and materials used

The data referring to the annual value of countries’ exports and imports in the period from 2000 to 2019 was obtained from Comtrade (United Nations, 2021) by 6-digits Harmonized System (HS), which were made available until February 2021. As some countries have not presented information about their trade flows for many years, we undertook the work of complementing this missing information. For this purpose, we initially mapped all the countries and years in which a lack of data was evident, either with regards to exports or imports. To cover these gaps, the global trade flows to these countries were identified and inverted. Thus, the export flows for the missing import data were inserted as imports for the countries in question and vice-versa. The values of these flows within the total amount of commercialized products vary between 1-5% of the total traded value, depending on the year. For some countries, in years when agri-trade information only is displayed by 4-digits HS instead of in 6-digit HS (as United Arab Emirates for example) we adjusted the products for this format. The USDA definition of agricultural products was used (United States, 2021).

We used the SQL Server to work in the databases and to estimate the indexes and Table Public to produce figures and the product described in Annex B. The figures presented throughout the text display parts of the results described and should be complemented by the figures shown in the link indicated in Annex B, which permit the online visualization of our results and general dataset in a more detailed manner, such as the identification of specific countries within each group, and the temporal evolution of the index. Those details could not be completely explored in this text due to spatial constraints.

In order to gain an overview of the structure of global agricultural trade, we initially illustrate the general political groupings and issue coalitions which we have chosen to treat within this field, as well as their relative size within agricultural trade flows. As can be seen in Figure 2, global agricultural export values are highly concentrated on the EU and United States (the Friends of Ambition Group or just USA & EU) and the Cairns Group, which together represent more than half of world export within this sector. For agricultural import value, the countries with no group and the G-33 are higher than Cairns Group. The G-10, and finally the G-90, stand as the least significant groupings in terms of global trade value, mainly in terms of exports.  

The G-33 and G-90 groups encompass the majority of the developing world, and have historically tended towards a defensive posture, motivated by food security-related concerns and the potentially negative impact of liberalizations (Costantini et al., 2007: 867). Other countries within the developing world can be found within the Cairns Group, which accounts for roughly 30% of global agricultural trade. Apart from agricultural exporters, such as Canada, Australia, and New Zeeland, the group comprises of countries within mainly Latin America and Southeast Asia. Historically, the Cairns Group has adopted a highly offensive stance within multilateral agricultural negotiations (Higgott and Cooper, 1990; Kaukab, 2007). Figure 2 also shows the evolution of agri-exports, which is marked by two different trends. The first period 2000-2011 presents a constant linear increase, which ends with the eruption of the crisis of 2008. In this period, the CAGR was 10%. In the second period from 2011-2019, the global growth of agri-trade was more limited, with a CAGR of 1%.           

More specifically, Figure 3 displays the imports and exports from the largest global agricultural markets, which encompass around 90% of the total agricultural trade value. This figure provides an overview of the different countries’ position within global agricultural trade in the first (2000) and last (2019) year of the period, showing their agricultural export and import values. The line dividing the graph represents the point at which total agricultural export values equal imports. Countries above this line thereby have an agricultural trade balance surplus, while countries below it present an agricultural trade balance deficit during the period analyzed. The value of the agricultural trade of countries such as the USA, the EU, and China (mainly in 2019) is much larger than that of many other countries. The Figure 1.B in Annex B permits a clearer visualization and identification of all the countries or select groups of countries (according to our definitions based on the WTO groups) as well as the annual values for each of these, hereby complementing Figure 3 which does not allow for a complete overview of the data analyzed.

In this study, we mapped the macro-trends and recent evolution of global agricultural trade from 2000-2019. For this  purpose, we assessed the distribution of global trade on countries and groupings, the current values of the agricultural trade balances, the sectorial and geographical diversification of each country’s agricultural imports and exports. Our findings were presented in schematic form, and their main implications highlighted within the analysis. In conclusion, we have chosen to stress certain elements within our findings, which, we believe, are highly noteworthy.    

(i) The impact of the rapidly increasing Chinese participation within global agricultural trade as well the increase in imports around the world has been highly significant to increase the exports during 2000-2011. Yet, as the Chinese imports have outpaced exports, resulting in a large agricultural trade balance deficit, countries exporting to China have been able to increase their shipments. This has made it possible for an agri-exporter such as Brazil to consistently raise its agricultural trade balance surplus over the period from 2000-2019.

(ii) For the period 2010-2019, growth in global agricultural trade has declined in the case of all WTO groups. Yet, this decline has been most accentuated in the case of the Cairns group in the case of export value, which together represent the largest share of global agricultural trade. For imports, G-10, G-90 and the countries without group presented a more accentuated decrease. This raises the question of whether global agribusiness trade has reached a level of structural saturation due to lower increases in demands for food imports, and in this case, which factors might be able to account for this development. In this regard, it is also worthwhile to consider the wave of sweeping multilateral trade liberalizations in the 1990s as well as its gradual stagnation in the 2000s, which could provide for an important point of departure for future studies to analyze this potential causal link. Possible political efforts to address food security concerns by increasing domestic supply in many countries should also be considered.   

(iii) Developing countries in the G-33 (with exception of China and India on the export) and G-90 are characterized by a very low degree of diversification of their agricultural trade. This concerns both sectorial and geographical diversification, in relation to both exports and imports. The low degree of sectorial and geographical export diversification may well mean that these countries become vulnerable to fluctuations within sectors and regional markets in which they are competitive. While our review of the evolution of sectorial export concentration of countries that present significant variations over the period from 2000-2019 points to a global trend of specialization in the export of agricultural commodities, developing countries would need to increase their sectorial diversification as well as geographic export diversification to benefit within this commercial environment. With regards to imports, the high degree of sectorial concentration characterizing groups of developing countries (mainly G-90 and G-33) is associated with a vulnerability to price shocks within specific markets and to sudden supply shortages, as it became evident during the two most recent food crises.

(iv) In contrast, the EU and the USA score high on both geographical and sectorial diversification on agribusiness trade flows (exports and imports). India and China also presented high diversification of exports (sectorial and geographic) and G-10 and the countries without group on sectorial diversification of imports. The high degree of diversification of agribusiness trade in the case of developed countries draws attention to the role which elevated per capita incomes are likely to play in driving demand for variated foodstuffs and agricultural products. This has provided market niches for some developing country exporters who have sought elevated price premiums within developed country markets. It will be important to observe to which extent similar trends become evident within rapidly growing developing markets, such as China, so that not only demand increases in terms of volumes, but also regarding diversification, can drive exports from other developing countries.       

(v) As the current study aims to provide a descriptive overview of structural trends within global agribusiness trade diversification and specialization, for future works, we recommend using the indices presented here as variables to compose an explanatory model of commercial trends.

The authors have not declared any conflict of interests.